The objective of the proposed work is to improve the treatment of epithelial malignancies with recombinant adenoviruses by restoring expression of the Coxsackie- and Adenovirus Receptor (CAR) at the surface of cancer cells. This molecule is the primary receptor for adenovirus and mediates initial attachment of the virus to target cells. Its expression is frequently reduced in cancer cells. The proposed project will explore the hypothesis that loss of CAR expression in cancer cells is the consequence of constitutive activation of the Ras/MAPK and TGF-beta signal transduction pathways resulting in repression of CAR gene transcription and alteration of the subcellular localization of CAR protein. To characterize alterations of the subcellular localization of CAR in response to Ras/MAPK and TGF-beta signaling, techniques for labeling surface proteins and cell fractionation will be utilized. This approach will be complemented by generating CAR-GFP fusion proteins, acting as reporter molecules for the assessment of alterations in CAR localization upon perturbation of signaling. Furthermore, changes of the composition of protein complexes containing CAR and concomitant modifications of the molecule will be assessed. To identify regulatory elements mediating CAR transcription in response to such signaling events, the regulatory region of the CAR gene will be defined. Luciferase-expressing reporter constructs containing the regulatory region will then be generated and, guided by computer-assisted sequence analysis, progressively mutated to identify relevant sequence elements. Binding of candidate transcription factors will be confirmed by mobility shift assays. The impact of Ras/MAPK and TGF-beta signaling will then be studied in a mouse model of skin cancer. Pharmacological inhibitors of both pathways will be administered and the resulting effect on CAR transcription and subcellular localization will be assessed. Finally, non-replicating recombinant adenoviruses expressing a reporter protein will be administered in conjunction with inhibitors of signaling to measure their impact on adenovirus entry into normal and cancer cells in vivo. Collectively, results from these studies will reveal detailed knowledge about the molecular link between CAR and the Ras/MAPK and TGF-beta signaling networks and explore the possibility of restoring CAR expression on cancer cells by pharmacological inhibition of these signaling circuits.